Predictive modeling in spine surgery

Utilization of Machine Learning to Model Important Features of 30-day Readmissions following Surgery for Metastatic Spinal Column Tumors: The Influence of Frailty

STUDY DESIGN

Retrospective cohort study.

OBJECTIVE

The aim of this study was to determine the relative importance and predicative power of the Hospital Frailty Risk Score (HFRS) on unplanned 30-day readmission after surgical intervention for metastatic spinal column tumors.

METHODS

All adult patients undergoing surgery for metastatic spinal column tumor were identified in the Nationwide Readmission Database from the years 2016 to 2018. Patients were categorized into 3 cohorts based on the criteria of the HFRS: Low(<5), Intermediate(5-14.9), and High(≥ 15). Random Forest (RF) classification was used to construct predictive models for 30-day patient readmission. Model performance was examined using the area under the receiver operating curve (AUC), and the Mean Decrease Gini (MDG) metric was used to quantify and rank features by relative importance.

RESULTS

There were 4346 patients included. The proportion of patients who required any readmission were higher among the Intermediate and High frailty cohorts when compared to the Low frailty cohort (Low:33.9% vs. Intermediate:39.3% vs. High:39.2%, P < .001). An RF classifier was trained to predict 30-day readmission on all features (AUC = .60) and architecturally equivalent model trained using only ten features with highest MDG (AUC = .59). Both models found frailty to have the highest importance in predicting risk of readmission. On multivariate regression analysis, Intermediate frailty [OR:1.32, CI(1.06,1.64), P = .012] was found to be an independent predictor of unplanned 30-day readmission.

CONCLUSION

Our study utilizes machine learning approaches and predictive modeling to identify frailty as a significant risk-factor that contributes to unplanned 30-day readmission after spine surgery for metastatic spinal column metastases.

Innovations in Spine Surgery: A Narrative Review of Current Integrative Technologies

Neurosurgical technologies have become increasingly more adaptive, featuring real-time and patient-specific guidance in preoperative, intraoperative, and postoperative settings. This review offers insight into how these integrative innovations compare with conventional approaches in spine surgery, focusing on machine learning (ML), artificial intelligence, augmented reality and virtual reality, and spinal navigation systems. Data on technology applications, diagnostic and procedural accuracy, intraoperative times, radiation exposures, postoperative outcomes, and costs were extracted and compared with conventional methods to assess their advantages and limitations. Preoperatively, augmented reality and virtual reality have applications in surgical training and planning that are more immersive, case specific, and risk-free and have been shown to enhance accuracy and reduce complications. ML algorithms have demonstrated high accuracy in predicting surgical candidacy (up to 92.1%) and tailoring personalized treatments based on patient-specific variables. Intraoperatively, advantages include more accurate pedicle screw insertion (96%-99% with ML), enhanced visualization, reduced radiation exposure (49 μSv with O-arm navigation vs. 556 μSv with fluoroscopy), increased efficiency, and potential for fewer intraoperative complications compared with conventional approaches. Postoperatively, certain ML and artificial intelligence models have outperformed conventional methods in predicting all postoperative complications of >6000 patients as well as predicting variables contributing to in-hospital and 90-day mortality. However, applying these technologies comes with limitations, such as longer operative times (up to 35.6% longer) with navigation, dependency on datasets, costs, accessibility, steep learning curve, and inherent software malfunctions. As these technologies advance, continuing to assess their efficacy and limitations will be crucial to their successful integration within spine surgery.

An Integrated Voice Recognition and Natural Language Processing Platform to Automatically Extract Thoracolumbar Injury Classification Score Features From Radiology Reports

BACKGROUND

Many predictive models for estimating clinical outcomes after spine surgery have been reported in the literature. However, implementation of predictive scores in practice is limited by the time-intensive nature of manually abstracting relevant predictors. In this study, we designed natural language processing (NLP) algorithms to automate data abstraction for the thoracolumbar injury classification score (TLICS).

METHODS

We retrieved the radiology reports of all Mayo Clinic patients with an International Classification of Diseases, 9th or 10th revision, code corresponding to a fracture of the thoracolumbar spine between January 2005 and October 2020. Annotated data were used to train an N-gram NLP model using machine learning methods, including random forest, stepwise linear discriminant analysis, k-nearest neighbors, and penalized logistic regression models.

RESULTS

A total of 1085 spine radiology reports were included in our analysis. Our dataset included 483 compression, 401 burst, 103 translational/rotational, and 98 distraction fractures. A total of 103 reports had documented an injury of the posterior ligamentous complex. The overall accuracy of the random forest model for fracture morphology feature detection was 76.96% versus 65.90% in the stepwise linear discriminant analysis, 50.69% in the k-nearest neighbors, and 62.67% in the penalized logistic regression. The overall accuracy to detect posterior ligamentous complex integrity was highest in the random forest model at 83.41%. Our random forest model was implemented in the backend of a web application in which users can dictate reports and have TLICS features automatically extracted.

CONCLUSIONS

We have developed a machine learning NLP model for extracting TLICS features from radiology reports, which we deployed in a web application that can be integrated into clinical practice.

Preserving privacy in big data research: the role of federated learning in spine surgery

PURPOSE

Integrating machine learning models into electronic medical record systems can greatly enhance decision-making, patient outcomes, and value-based care in healthcare systems. Challenges related to data accessibility, privacy, and sharing can impede the development and deployment of effective predictive models in spine surgery. Federated learning (FL) offers a decentralized approach to machine learning that allows local model training while preserving data privacy, making it well-suited for healthcare settings. Our objective was to describe federated learning solutions for enhanced predictive modeling in spine surgery.

METHODS

The authors reviewed the literature.

RESULTS

FL has promising applications in spine surgery, including telesurgery, AI-based prediction models, and medical image segmentation. Implementing FL requires careful consideration of infrastructure, data quality, and standardization, but it holds the potential to revolutionize orthopedic surgery while ensuring patient privacy and data control.

CONCLUSIONS

Federated learning shows great promise in revolutionizing predictive modeling in spine surgery by addressing the challenges of data privacy, accessibility, and sharing. The applications of FL in telesurgery, AI-based predictive models, and medical image segmentation have demonstrated their potential to enhance patient outcomes and value-based care.

Predictive Algorithm for Surgery Recommendation in Thoracolumbar Burst Fractures Without Neurological Deficits

STUDY DESIGN

Predictive algorithm via decision tree.

OBJECTIVES

Artificial intelligence (AI) remain an emerging field and have not previously been used to guide therapeutic decision making in thoracolumbar burst fractures. Building such models may reduce the variability in treatment recommendations. The goal of this study was to build a mathematical prediction rule based upon radiographic variables to guide treatment decisions.

METHODS

Twenty-two surgeons from the AO Knowledge Forum Trauma reviewed 183 cases from the Spine TL A3/A4 prospective study (classification, degree of certainty of posterior ligamentous complex (PLC) injury, use of M1 modifier, degree of comminution, treatment recommendation). Reviewers' regions were classified as Europe, North/South America and Asia. Classification and regression trees were used to create models that would predict the treatment recommendation based upon radiographic variables. We applied the decision tree model which accounts for the possibility of non-normal distributions of data. Cross-validation technique as used to validate the multivariable analyses.

RESULTS

The accuracy of the model was excellent at 82.4%. Variables included in the algorithm were certainty of PLC injury (%), degree of comminution (%), the use of M1 modifier and geographical regions. The algorithm showed that if a patient has a certainty of PLC injury over 57.5%, then there is a 97.0% chance of receiving surgery. If certainty of PLC injury was low and comminution was above 37.5%, a patient had 74.2% chance of receiving surgery in Europe and Asia vs 22.7% chance in North/South America. Throughout the algorithm, the use of the M1 modifier increased the probability of receiving surgery by 21.4% on average.

CONCLUSION

This study presents a predictive analytic algorithm to guide decision-making in the treatment of thoracolumbar burst fractures without neurological deficits. PLC injury assessment over 57.5% was highly predictive of receiving surgery (97.0%). A high degree of comminution resulted in a higher chance of receiving surgery in Europe or Asia vs North/South America. Future studies could include clinical and other variables to enhance predictive ability or use machine learning for outcomes prediction in thoracolumbar burst fractures.

Artificial intelligence-based analysis of associations between learning curve and clinical outcomes in endoscopic and microsurgical lumbar decompression surgery

PURPOSE

A common spine surgery procedure involves decompression of the lumbar spine. The impact of the surgeon's learning curve on relevant clinical outcomes is currently not well examined in the literature. A variety of machine learning algorithms have been investigated in this study to determine how a surgeon's learning curve and other clinical parameters will influence prolonged lengths of stay (LOS), extended operating times (OT), and complications, as well as whether these clinical parameters can be reliably predicted.

METHODS

A retrospective monocentric cohort study of patients with lumbar spinal stenosis treated with microsurgical (MSD) and full-endoscopic (FED) decompression was conducted. The study included 206 patients with lumbar spinal stenosis who underwent FED (63; 30.6%) and MSD (118; 57.3%). Prolonged LOS and OT were defined as those exceeding the 75th percentile of the cohort. Furthermore, complications were assessed as a dependent variable. Using unsupervised learning, clusters were identified in the data, which helped distinguish between the early learning curve (ELC) and the late learning curve (LLC). From 15 algorithms, the top five algorithms that best fit the data were selected for each prediction task. We calculated the accuracy of prediction (Acc) and the area under the curve (AUC). The most significant predictors were determined using a feature importance analysis.

RESULTS

For the FED group, the median number of surgeries with case surgery type at the time of surgery was 72 in the ELC group and 274 in the LLC group. FED patients did not significantly differ in outcome variables (LOS, OT, complication rate) between the ELC and LLC group. The random forest model demonstrated the highest mean accuracy and AUC across all folds for each classification task. For OT, it achieved an accuracy of 76.08% and an AUC of 0.89. For LOS, the model reached an accuracy of 83.83% and an AUC of 0.91. Lastly, in predicting complications, the random forest model attained the highest accuracy of 89.90% and an AUC of 0.94. Feature importance analysis indicated that LOS, OT, and complications were more significantly affected by patient characteristics than the surgical technique (FED versus MSD) or the surgeon's learning curve.

CONCLUSIONS

A median of 72 cases of FED surgeries led to comparable clinical outcomes in the early learning curve phase compared to experienced surgeons. These outcomes seem to be more significantly affected by patient characteristics than the learning curve or the surgical technique. Several study variables, including the learning curve, can be used to predict whether lumbar decompression surgery will result in an increased LOS, OT, or complications. To introduce the provided prediction tools into clinics, the algorithms need to be implemented into open-source software and externally validated through large-scale randomized controlled trials.

An Artificial Intelligence Approach to Predicting Unplanned Intubation Following Anterior Cervical Discectomy and Fusion

STUDY DESIGN

Level III retrospective database study.

OBJECTIVES

The purpose of this study is to determine if machine learning algorithms are effective in predicting unplanned intubation following anterior cervical discectomy and fusion (ACDF).

METHODS

The National Surgical Quality Initiative Program (NSQIP) was queried to select patients who had undergone ACDF. Machine learning analysis was conducted in Python and multivariate regression analysis was conducted in R. C-Statistics area under the curve (AUC) and prediction accuracy were used to measure the classifier's effectiveness in distinguishing cases.

RESULTS

In total, 54 502 patients met the study criteria. Of these patients, .51% underwent an unplanned re-intubation. Machine learning algorithms accurately classified between 72%-100% of the test cases with AUC values of between .52-.77. Multivariable regression indicated that the number of levels fused, male sex, COPD, American Society of Anesthesiologists (ASA) > 2, increased operating time, Age > 65, pre-operative weight loss, dialysis, and disseminated cancer were associated with increased risk of unplanned intubation.

CONCLUSIONS

The models presented here achieved high accuracy in predicting risk factors for re-intubation following ACDF surgery. Machine learning analysis may be useful in identifying patients who are at a higher risk of unplanned post-operative re-intubation and their treatment plans can be modified to prophylactically prevent respiratory compromise and consequently unplanned re-intubation.

Revolutionizing Spinal Care: Current Applications and Future Directions of Artificial Intelligence and Machine Learning

Artificial intelligence (AI) and machine learning (ML) are rapidly becoming integral components of modern healthcare, offering new avenues for diagnosis, treatment, and outcome prediction. This review explores their current applications and potential future in the field of spinal care. From enhancing imaging techniques to predicting patient outcomes, AI and ML are revolutionizing the way we approach spinal diseases. AI and ML have significantly improved spinal imaging by augmenting detection and classification capabilities, thereby boosting diagnostic accuracy. Predictive models have also been developed to guide treatment plans and foresee patient outcomes, driving a shift towards more personalized care. Looking towards the future, we envision AI and ML further ingraining themselves in spinal care with the development of algorithms capable of deciphering complex spinal pathologies to aid decision making. Despite the promise these technologies hold, their integration into clinical practice is not without challenges. Data quality, integration hurdles, data security, and ethical considerations are some of the key areas that need to be addressed for their successful and responsible implementation. In conclusion, AI and ML represent potent tools for transforming spinal care. Thoughtful and balanced integration of these technologies, guided by ethical considerations, can lead to significant advancements, ushering in an era of more personalized, effective, and efficient healthcare.

Clinical outcomes and prediction nomogram model for postoperative hemoglobin < 80 g/L in patients following primary lumbar interbody fusion surgery

OBJECTIVE

To analyze the association between different postoperative hemoglobin (Hb) levels and postoperative outcomes in patients who have undergone primary lumbar interbody fusion, and to investigate the risk factors and establish a predictive nomogram mode for postoperative Hb < 80 g/L.

METHODS

We retrospectively analyzed 726 cases who underwent primary lumbar interbody fusion surgery between January 2018 and December 2021in our hospital. All patients were divided into three groups according to the postoperative Hb levels (< 70 g/L, 70-79 g/L, ≥ 80 g/L). The postoperative outcomes among the three groups were compared, and the risk factors for postoperative Hb < 80 g/L were identified by univariate and multivariable logistic regression analysis. Based on these independent predictors, a nomogram model was developed. Predictive discriminative and accuracy ability of the predicting model was assessed using the concordance index (C-index) and calibration plot. Clinical application was validated using decision curve analysis. Internal validation was performed using the bootstrapping validation.

RESULTS

Patients with postoperative Hb < 80 g/L had higher rates of postoperative blood transfusion, a greater length of stay, higher rates of wound complications, and higher hospitalization costs than those with postoperative Hb ≥ 80 g/L. Preoperative Hb, preoperative platelets, fusion segments, body mass index, operation time, and intraoperative blood loss independently were associated with postoperative Hb < 80 g/L. Intraoperative blood salvage was found to be a negative predictor for postoperative Hb < 80 g/L (OR, 0.21 [95% CI 0.09-0.50]). The area under the curve of the nomogram model was 0.950. After internal validations, the C-index of the model was 0.939. The DCA and calibration curve suggested that the nomogram model had a good consistency and clinical utility.

CONCLUSIONS

Postoperative Hb < 80 g/L in patients following primary lumbar interbody fusion surgery increased blood transfusions requirement and was independently associated with poor outcomes. A novel nomogram model was established and could conveniently predict the risk of postoperative Hb < 80 g/L in patients after this type of surgery.

Artificial Intelligence in Modern Orthopaedics: Current and Future Applications

➢

With increasing computing power, artificial intelligence (AI) has gained traction in all aspects of health care delivery. Orthopaedics is no exception because the influence of AI technology has become intricately linked with its advancement as evidenced by increasing interest and research.

➢

This review is written for the orthopaedic surgeon to develop a better understanding of the main clinical applications and potential benefits of AI within their day-to-day practice.

➢

A brief and easy-to-understand foundation for what AI is and the different terminology used within the literature is first provided, followed by a summary of the newest research on AI applications demonstrating increased accuracy and convenience in risk stratification, clinical decision-making support, and robotically assisted surgery.

Performance of Artificial Intelligence-Based Algorithms to Predict Prolonged Length of Stay after Lumbar Decompression Surgery

Background: Decompression of the lumbar spine is one of the most common procedures performed in spine surgery. Hospital length of stay (LOS) is a clinically relevant metric used to assess surgical success, patient outcomes, and socioeconomic impact. This study aimed to investigate a variety of machine learning and deep learning algorithms to reliably predict whether a patient undergoing decompression of lumbar spinal stenosis will experience a prolonged LOS. Methods: Patients undergoing treatment for lumbar spinal stenosis with microsurgical and full-endoscopic decompression were selected within this retrospective monocentric cohort study. Prolonged LOS was defined as an LOS greater than or equal to the 75th percentile of the cohort (normal versus prolonged stay; binary classification task). Unsupervised learning with K-means clustering was used to find clusters in the data. Hospital stay classes were predicted with logistic regression, RandomForest classifier, stochastic gradient descent (SGD) classifier, K-nearest neighbors, Decision Tree classifier, Gaussian Naive Bayes (GaussianNB), support vector machines (SVM), a custom-made convolutional neural network (CNN), multilayer perceptron artificial neural network (MLP), and radial basis function neural network (RBNN) in Python. Prediction accuracy and area under the curve (AUC) were calculated. Feature importance analysis was utilized to find the most important predictors. Further, we developed a decision tree based on the Chi-square automatic interaction detection (CHAID) algorithm to investigate cut-offs of predictors for clinical decision-making. Results: 236 patients and 14 feature variables were included. K-means clustering separated data into two clusters distinguishing the data into two patient risk characteristic groups. The algorithms reached AUCs between 67.5% and 87.3% for the classification of LOS classes. Feature importance analysis of deep learning algorithms indicated that operation time was the most important feature in predicting LOS. A decision tree based on CHAID could predict 84.7% of the cases. Conclusions: Machine learning and deep learning algorithms can predict whether patients will experience an increased LOS following lumbar decompression surgery. Therefore, medical resources can be more appropriately allocated to patients who are at risk of prolonged LOS.

Improving Surgical Triage in Spine Clinic: Predicting Likelihood of Surgery Using Machine Learning

BACKGROUND

Correctly triaging patients to a surgeon or a nonoperative provider is an important part of the referral process. Clinics typically triage new patients based on simple intake questions. This is time-consuming and does not incorporate objective data. Our goal was to use machine learning to more accurately screen surgical candidates seen in a spine clinic.

METHODS

Using questionnaire data and magnetic resonance imaging reports, a set of artificial neural networks was trained to predict whether a patient would be recommended for spine surgery. Questionnaire responses included demographics, chief complaint, and pain characteristics. The primary end point was the surgeon's determination of whether a patient was an operative candidate. Model accuracy in predicting this end point was assessed using a separate subset of patients.

RESULTS

The retrospective dataset included 1663 patients in cervical and lumbar cohorts. Questionnaire data were available for all participants, and magnetic resonance imaging reports were available for 242 patients. Within 6 months of initial evaluation, 717 (43.1%) patients were deemed surgical candidates by the surgeon. Our models predicted surgeons' recommendations with area under the curve scores of 0.686 for lumbar (positive predictive value 66%, negative predictive value 80%) and 0.821 for cervical (positive predictive value 83%, negative predictive value 85%) patients.

CONCLUSIONS

Our models used patient data to accurately predict whether patients will receive a surgical recommendation. The high negative predictive value demonstrates that this approach can reduce the burden of nonsurgical patients in surgery clinic without losing many surgical candidates. This could reduce unnecessary visits for patients, increase the proportion of operative candidates seen by surgeons, and improve quality of patient care.

Feasibility and Assessment of a Machine Learning-Based Predictive Model of Outcome After Lumbar Decompression Surgery

STUDY DESIGN

Retrospective study at a unique center.

OBJECTIVE

The aim of this study is twofold, to develop a virtual patients model for lumbar decompression surgery and to evaluate the precision of an artificial neural network (ANN) model designed to accurately predict the clinical outcomes of lumbar decompression surgery.

METHODS

We performed a retrospective study of complete Electronic Health Records (EHR) to identify potential unfavorable criteria for spine surgery (predictors). A cohort of synthetics EHR was created to classify patients by surgical success (green zone) or partial failure (orange zone) using an Artificial Neural Network which screens all the available predictors.

RESULTS

In the actual cohort, we included 60 patients, with complete EHR allowing efficient analysis, 26 patients were in the orange zone (43.4%) and 34 were in the green zone (56.6%). The average positive criteria amount for actual patients was 8.62 for the green zone (SD+/- 3.09) and 10.92 for the orange zone (SD 3.38). The classifier (a neural network) was trained using 10,000 virtual patients and 2000 virtual patients were used for test purposes. The 12,000 virtual patients were generated from the 60 EHR, of which half were in the green zone and half in the orange zone. The model showed an accuracy of 72% and a ROC score of 0.78. The sensitivity was 0.885 and the specificity 0.59.

CONCLUSION

Our method can be used to predict a favorable patient to have lumbar decompression surgery. However, there is still a need to further develop its ability to analyze patients in the "failure of treatment" zone to offer precise management of patient health before spinal surgery.

Machine Learning Applications of Surgical Imaging for the Diagnosis and Treatment of Spine Disorders: Current State of the Art

Recent developments in machine learning (ML) methods demonstrate unparalleled potential for application in the spine. The ability for ML to provide diagnostic faculty, produce novel insights from existing capabilities, and augment or accelerate elements of surgical planning and decision making at levels equivalent or superior to humans will tremendously benefit spine surgeons and patients alike. In this review, we aim to provide a clinically relevant outline of ML-based technology in the contexts of spinal deformity, degeneration, and trauma, as well as an overview of commercial-level and precommercial-level surgical assist systems and decisional support tools. Furthermore, we briefly discuss potential applications of generative networks before highlighting some of the limitations of ML applications. We conclude that ML in spine imaging represents a significant addition to the neurosurgeon's armamentarium-it has the capacity to directly address and manifest clinical needs and improve diagnostic and procedural quality and safety-but is yet subject to challenges that must be addressed before widespread implementation.

Machine Learning Prediction of Length of Stay in Adult Spinal Deformity Patients Undergoing Posterior Spine Fusion Surgery

(1) Background: Length of stay (LOS) is a commonly reported metric used to assess surgical success, patient outcomes, and economic impact. The focus of this study is to use a variety of machine learning algorithms to reliably predict whether a patient undergoing posterior spinal fusion surgery treatment for Adult Spine Deformity (ASD) will experience a prolonged LOS. (2) Methods: Patients undergoing treatment for ASD with posterior spinal fusion surgery were selected from the American College of Surgeon's NSQIP dataset. Prolonged LOS was defined as a LOS greater than or equal to 9 days. Data was analyzed with the Logistic Regression, Decision Tree, Random Forest, XGBoost, and Gradient Boosting functions in Python with the Sci-Kit learn package. Prediction accuracy and area under the curve (AUC) were calculated. (3) Results: 1281 posterior patients were analyzed. The five algorithms had prediction accuracies between 68% and 83% for posterior cases (AUC: 0.566-0.821). Multivariable regression indicated that increased Work Relative Value Units (RVU), elevated American Society of Anesthesiologists (ASA) class, and longer operating times were linked to longer LOS. (4) Conclusions: Machine learning algorithms can predict if patients will experience an increased LOS following ASD surgery. Therefore, medical resources can be more appropriately allocated towards patients who are at risk of prolonged LOS.

Inclusion of Frailty Improves Predictive Modeling for Postoperative Outcomes in Surgical Management of Primary and Secondary Lumbar Spine Tumors

BACKGROUND

Malignant spinal tumors are common, continually increasing in incidence as a function of improved survival times for patients with cancer. Using predictive analytics and propensity score matching, we evaluated the influence of frailty on postoperative complications compared with age in patients with malignant neoplasms of the lumbar spine.

METHODS

We used the Nationwide Readmissions Database from 2016 and 2017 to identify patients with malignant neoplasms of the lumbar spine who received a fusion procedure. Patient frailty was queried using the Johns Hopkins Adjusted Clinical Groups. Propensity score matching for age, sex, Charlson Comorbidity Index, surgical approach, and number of levels fused was implemented between frail and nonfrail patients, identifying 533 frail patients and 538 nonfrail patients. The area under the curve (AUC) of each ROC served as a proxy for model performance.

RESULTS

Frail patients reported significantly higher inpatient lengths of stay, costs, infection, posthemorrhagic anemia, and urinary tract infections (P < 0.05). In addition, frail patients were more often discharged to skilled nursing facilities and short-term hospitals compared with nonfrail patients (P < 0.0001). Regression models for mortality (AUC = 0.644), nonroutine discharge (AUC = 0.600), and acute infection (AUC = 0.666) were improved when using frailty as the primary predictor. These models were also improved using frailty when predicting 30-day readmission and 90-day hardware failure.

CONCLUSIONS

Frailty demonstrated a significant relationship with increased postoperative patient complications, length of stay, costs, and acute complications in patients receiving fusion following resection of a malignant neoplasm of the lumbar spine region. Frailty demonstrated better predictive validity of outcomes compared with patient age.

Predicting Readmission After Anterior, Posterior, and Posterior Interbody Lumbar Spinal Fusion: A Neural Network Machine Learning Approach

BACKGROUND

Readmission after spine surgery is costly and a relatively common occurrence. Previous research identified several risk factors for readmission; however, the conclusions remain equivocal. Machine learning algorithms offer a unique perspective in analysis of risk factors for readmission and can help predict the likelihood of this occurrence. This study evaluated a neural network (NN), a supervised machine learning technique, to determine whether it could predict readmission after 3 lumbar fusion procedures.

METHODS

The American College of Surgeons National Surgical Quality Improvement Program database was queried between 2009 and 2018. Patients who had undergone anterior, lateral, and/or posterior lumbar fusion were included in the study. The Python scikit Learn package was used to run the NN algorithm. A multivariate regression was performed to determine risk factors for readmission.

RESULTS

There were 63,533 patients analyzed (12,915 anterior lumbar interbody fusion, 27,212 posterior lumbar interbody fusion, and 23,406 posterior spinal fusion cases). The NN algorithm was able to successfully predict 30-day readmission for 94.6% of anterior lumbar interbody fusion, 94.0% of posterior lumbar interbody fusion, and 92.6% of posterior spinal fusion cases with area under the curve values of 0.64-0.65. Multivariate regression indicated that age >65 years and American Society of Anesthesiologists class >II were linked to increased risk for readmission for all 3 procedures.

CONCLUSIONS

The accurate metrics presented indicate the capability for NN algorithms to predict readmission after lumbar arthrodesis. Moreover, the results of this study serve as a catalyst for further research into the utility of machine learning in spine surgery.

Establishment and assessment of a nomogram for predicting blood transfusion risk in posterior lumbar spinal fusion

Background

The aim of this study was to determine the risk factors and develop a nomogram for blood transfusions after posterior lumbar spinal fusion (PSL).

Methods

We conducted a retrospective, single-center study based on 885 patients receiving PSL, and data was obtained from May 2015 to September 2019. Univariable and multivariable logistics regression analysis were conducted to identify risk factors for blood transfusion, and a nomogram was constructed to individually evaluate the risk of blood transfusion. Discrimination, calibration, and clinical usefulness were validated by the receiver operating characteristics (ROC), C-index, calibration plot, and decision curve analysis, respectively. Bootstrapping validation was performed to assess the performance of the model.

Results

Of 885 patients, 885 were enrolled in the final study population, and 289 received blood transfusion. Statistical analyses showed that low preoperative hemoglobin (Hb), longer time to surgery, operative time, levels of fusion > 1, longer surgery duration, and higher total intraoperative blood loss (IBL) were the risk factors for transfusion. The C-index was 0.898 (95% CI 0.847–0.949) in this dataset and 0.895 in bootstrapping validation, respectively. Calibration curve showed satisfied discrimination and calibration of the nomogram. Decision curve analysis (DCA) shown that the nomogram was clinical utility.

Conclusions

In summary, we investigated the relationship between the blood transfusion requirement and predictors: levels of fusion, operative time, time to surgery, total intraoperative EBL, and preoperative Hb level. Our nomogram with a robust performance in the assessment of risk of transfusion can contribute to clinicians in making clinical decision. However, external validation is still needed in the further.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-020-02053-2.

Predicting Postoperative Mortality After Metastatic Intraspinal Neoplasm Excision: Development of a Machine-Learning Approach

OBJECTIVE

Mortality following surgical resection of spinal tumors is a devastating outcome. Naïve Bayes machine learning algorithms may be leveraged in surgical planning to predict mortality. In this investigation, we use a Naïve Bayes classification algorithm to predict mortality following spinal tumor excision within 30 days of surgery.

METHODS

Patients who underwent laminectomies between 2006 and 2018 for excisions of intraspinal neoplasms were selected from the National Surgical Quality Initiative Program. Naïve Bayes classifier analysis was conducted in Python. The area under the receiver operating curve (AUC) was calculated to evaluate the classifier's ability to predict mortality within 30 days of surgery. Multivariable logistic regression analysis was performed in R to identify risk factors for 30-day postoperative mortality.

RESULTS

In total, 2094 spine tumor surgery patients were included in the study. The 30-day mortality rate was 5.16%. The classifier yielded an AUC of 0.898, which exceeds the predictive capacity of the National Surgical Quality Initiative Program mortality probability calculator's AUC of 0.722 (P < 0.0001). The multivariable regression indicated that smoking history, chronic obstructive pulmonary disease, disseminated cancer, bleeding disorder history, dyspnea, and low albumin levels were strongly associated with 30-day mortality.

CONCLUSIONS

The Naïve Bayes classifier may be used to predict 30-day mortality for patients undergoing spine tumor excisions, with an increasing degree of accuracy as the model better performs by learning continuously from the input patient data. Patient outcomes can be improved by identifying high-risk populations early using the algorithm and applying that data to inform preoperative decision making, as well as patient selection and education.

Predictive Model for Selection of Upper Treated Vertebra Using a Machine Learning Approach

OBJECTIVE

To train and validate an algorithm mimicking decision making of experienced surgeons regarding upper instrumented vertebra (UIV) selection in surgical correction of thoracolumbar adult spinal deformity.

METHODS

A retrospective review was conducted of patients with adult spinal deformity who underwent fusion of at least the lumbar spine (UIV > L1 to pelvis) during 2013-2018. Demographic and radiographic data were collected. The sample was stratified into 3 groups: training (70%), validation (15%) and performance testing (15%). Using a deep learning algorithm, a neural network model was trained to select between upper thoracic (T1-T6) and lower thoracic (T7-T12) UIV. Parameters used in the deep learning algorithm included demographics, coronal and sagittal preoperative alignment, and postoperative pelvic incidence-lumbar lordosis mismatch.

RESULTS

The study included 143 patients (mean age 63.3 ± 10.6 years, 81.8% women) with moderate to severe deformity (maximum Cobb angle: 43° ± 22°; T1 pelvic angle: 27° ± 14°; pelvic incidence-lumbar lordosis mismatch: 22° ± 21°). Patients underwent a significant change in lumbar alignment (Δpelvic incidence-lumbar lordosis mismatch: 21° ± 16°, P < 0.001); 35.0% had UIV in the upper thoracic region, and 65.0% had UIV in the lower thoracic region. At 1 year, revision rate was 11.9%, and rate of radiographic proximal junctional kyphosis was 29.4%. Neural network comprised 8 inputs, 10 hidden neurons, and 1 output (upper thoracic or lower thoracic). After training, results demonstrated an accuracy of 81.0%, precision of 87.5%, and recall of 87.5% on testing.

CONCLUSIONS

An artificial neural network successfully mimicked 2 lead surgeons' decision making in the selection of UIV for adult spinal deformity correction. Future models integrating surgical outcomes should be developed.

A predictive-modeling based screening tool for prolonged opioid use after surgical management of low back and lower extremity pain

BACKGROUND CONTEXT

Outpatient postoperative pain management in spine patients, specifically involving the use of opioids, demonstrates significant variability.

PURPOSE

Using preoperative risk factors and 30-day postoperative opioid prescribing patterns, we developed models for predicting long-term opioid use in patients after elective spine surgery.

STUDY DESIGN/SETTING

This retrospective cohort study utilizes inpatient, outpatient, and pharmaceutical data from MarketScan databases (Truven Health).

PATIENT SAMPLE

In all, 19,317 patients who were newly diagnosed with low back or lower extremity pain (LBP or LEP) between 2008 and 2015 and underwent thoracic or lumbar surgery within 1 year after diagnosis were enrolled. Some patients initiated opioids after diagnosis but all patients were opioid-naïve before the diagnosis.

OUTCOME MEASURES

Long-term opioid use was defined as filling ≥180 days of opioids within one year after surgery.

METHODS

Using demographic variables, medical and psychiatric comorbidities, preoperative opioid use, and 30-day postoperative opioid use, we generated seven models on 80% of the dataset and tested the models on the remaining 20%. We used three regression-based models (full logistic regression, stepwise logistic regression, least absolute shrinkage and selection operator), support vector machine, two tree-based models (random forest, stochastic gradient boosting), and time-varying convolutional neural network. Area under the curve (AUC), Brier index, sensitivity, and calibration curves were used to assess the discrimination and calibration of the models.

RESULTS

We identified 903 (4.6%) of patients who met criteria for long-term opioid use. The regression-based models demonstrated the highest AUC, ranging from 0.835 to 0.847, and relatively high sensitivities, predicting between 74.9% and 76.5% of the long-term opioid use patients in the test dataset. The three strongest positive predictors of long-term opioid use were high preoperative opioid use (OR 2.70; 95% confidence interval [CI] 2.27-3.22), number of days with active opioid prescription between postoperative days 15 to 30 (OR 1.10; 95%CI 1.07-1.12), and number of dosage increases between postoperative day 15 to 30 (OR 1.71, 95%CI 1.41-2.08). The strongest negative predictors were number of dosage decreases in the 30-day postoperative period.

CONCLUSIONS

We evaluated several predictive models for postoperative long-term opioid use in a large cohort of patients with LBP or LEP who underwent surgery. A regression-based model with high sensitivity and AUC is provided online to screen patients for high risk of long-term opioid use based on preoperative risk factors and opioid prescription patterns in the first 30 days after surgery. It is hoped that this work will improve identification of patients at high risk of prolonged opioid use and enable early intervention and counseling.

Predicting Patient-Centered Outcomes from Spine Surgery Using Risk Assessment Tools: a Systematic Review

PURPOSE OF REVIEW

The purpose of this systematic review is to evaluate the current literature in patients undergoing spine surgery in the cervical, thoracic, and lumbar spine to determine the available risk assessment tools to predict the patient-centered outcomes of pain, disability, physical function, quality of life, psychological disposition, and return to work after surgery.

RECENT FINDINGS

Risk assessment tools can assist surgeons and other healthcare providers in identifying the benefit-risk ratio of surgical candidates. These tools gather demographic, medical history, and other pertinent patient-reported measures to calculate a probability utilizing regression or machine learning statistical foundations. Currently, much is still unknown about the use of these tools to predict quality of life, disability, and other factors following spine surgery. A systematic review was conducted using PRISMA guidelines that identified risk assessment tools that utilized patient-reported outcome measures as part of the calculation. From 8128 identified studies, 13 articles met inclusion criteria and were accepted into this review. The range of c-index values reported in the studies was between 0.63 and 0.84, indicating fair to excellent model performance. Post-surgical patient-reported outcomes were identified in the following categories (n = total number of predictive models): return to work (n = 3), pain (n = 9), physical functioning and disability (n = 5), quality of life (QOL) (n = 6), and psychosocial disposition (n = 2). Our review has synthesized the available evidence on risk assessment tools for predicting patient-centered outcomes in patients undergoing spine surgery and described their findings and clinical utility.